基于多机编队有源组网探测的多目标跟踪算法  

A Multi-target Tracking Algorithm based on Networking Detection of Multiple Airplanes

作　　者：黄大羽[1,2] 倪忠建[1,2] 徐飞[1,2] 

机构地区：[1]中国航空无线电电子研究所,上海200233 [2]航空电子系统综合技术重点实验室,上海200233

出　　处：《探测与控制学报》2015年第4期46-52,共7页Journal of Detection & Control

基　　金：国家级重点实验室基金项目资助(基金号不公开)

摘　　要：针对空战背景下低可观测目标的跟踪问题,提出了基于多机编队有源组网探测的多目标跟踪算法。该算法根据多平台信息分配模型与评分,设计基于评分的求解方法选取可行分配模式,合成等效测量;结合多假设思想,以航迹树的形式组织可能目标,通过设计航迹树的分支与删除策略,实现虚警的剔除以及航迹的起始、维持与终结,通过设计航迹树显示逻辑,显示当前时刻目标的航迹批号与运动参数。仿真结果表明,提出的算法能够在低可观测条件下有效并稳定地跟踪多个目标。For multiple targets tracking under the condition of low detection probability, a new multi-target tracking algorithm based on networking airplanes was proposed. A second-best technique to obtain equivalent measures according to dynamic allocation model and scoring functions were designed. Then the algorithm re- moved the false alarms, initialized the tracks, maintained the tracks and terminated the tracks by multiple hy- pothesis testing. Meanwhile, display logic was applied to show the state estimation and the hatch of targets at each scanning. Simulation results showed that the proposed algorithm could efficiently and steadily track multi- pie targets under the condition of low detection probability.

关 键 词：多目标跟踪 多机编队 多假设 低可观测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]